Forum Moderators: martinibuster
Fortunately I was able to compare the stats across several web sites and the 33% was very consistent on the larger sites however on some smaller sites the bot ratio climbed to 50%!
Are these stats the norm?
Has anyone else used Adsense and compared against web logs etc and have any firm stats?
Per AWStats, I had a ratio of bots:nots of 1:21, 5% here. I can't vouch for how accurate AWStats bot fingerprinting is, but their "pages viewed" number seems to line up quite well with my imps as reported by the stats page, so I have some faith in it.
More than 75% of my 'users' are bots, not humans. It will depend on your site size and conditions a lot.
If you have a 10-page site that is ranked well in Google for a high-traffic keyword, then the odds are that bots will be only a tiny tiny fraction of your hit total. After all, Gbot et al. only have 10 pages to spider. For a large site like Webmasterworld, except imagine one without quite the dedicated user base, and you can easily get Gbot, Slurp, et al. to spider thousands upon thousands of pages each per day. So you might find 100,000 "users" from pure bots.
I reckon that something like 1/3 to 1/2 of visits to my site(s) are bots, both legit and scum...
And because of the scumbots I have to work quite hard to prevent the sites and my connections from being taken down by them (and also end up putting limits on legit users). Depressing sometimes when you are giving something away for free and people still try so hard to steal it anyway that the whole service is brought to its knees...
On the other hand, in the old days when G was the new kid on the block, G's bots crippled my connections and so even G has to be kept in check sometimes! B^>
Rgds
Damon
Sure, you can get more sophisticated in your analysis of web users than Adsense stats is able to provide. But if you are starting from the premise that your stats don't differentiate known bots from humans, then by all means, Adsense is much much more accurate.
The reason I've started to analyse them more closely is that until recently all our simple stats were provided by Webalizer however I recently moved some sites to a server which provides Awstats and the difference in visitor and page numbers amazed me.
This is why I then decided to do a much more detailed analysis comparing my Adsense stats and impressions to those I had been using and these were the results I came up with.
One other strange phenomenon I've noticed this last few weekends has been much less bot activity distorting visitor figures.
Is this actually happening or is it purely conincidental?
